欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
基于粗糙模糊集的规则提取方法通常分为两步：首先利用粗糙模糊集进行属性约简，然后采用提取模糊规则的方法提取规则．在规则提取的预处理阶段通过属性约简某种程度上可以缩短规则提取的时间，但其固有的不足导致不利于产生良好的规则．在模糊规则产生过程中避开属性约简，可以提高规则提取方法的适用性，降低计算复杂度．本文提出了动态粗糙模糊集的概念，基于此的规则提取算法不再依赖于属性约简，而是基于粒度序和逐步缩小的论域．首先，通过两种不同方式定义了动态粗糙模糊集并得到一些重要性质；在此基础上提出一种新的模糊规则提取算法；最后通过对比实验说明了算法的有效性．
英文摘要：
Most rule induction algorithms based on rough fuzzy sets theory often include two steps： attributes reduction based on rough fuzzy sets and fuzzy rules induction based on conventional rules mining algorithms. It＇s s useful to shorten time to some extent by attributes reduction in preprocessor of rule mining. However, attributes reduction may make against the induction of fine rules due to the flaws of itself. Avoiding the process of attributes reduction in fuzzy rules induction permits to improve the adaptability of generating fuzzy rules and reduce computational complexity. In this paper, the dynamic rough fuzzy sets were presented. A rule induction algorithm which, different from most known fuzzy rules induction , is not based on attributes reduction but granulation order and dwindle universe was designed. Firstly, the dynamic rough fuzzy sets were defined in two different ways and some important properties were obtained, creating a base for induction of fuzzy rules. Secondly, an algorithm, based on dynamic rough fuzzy sets was put forward for decision rule mining. At last, application of the algorithm was illustrated by an example. The results showed that the algorithm was effeetive, supported by comparisons to the application of fuzzy rules induction based on attributes reduction.
同期刊论文项目
同项目期刊论文
期刊信息
位置：